User

Difference between revisions of "Ana Calarasanu"

From LRDE

Line 15: Line 15:
   
 
==Projects==
 
==Projects==
  +
[https://www.lrde.epita.fr/wiki/Evaltex '''EvaLTex'''] (Evaluating Text Localization) is a unified evaluation framework used to measure the performance of text detection and text segmentation algorithms. It takes as input text objects represented either by rectangle coordinates or by irregular masks. The output consists of a set of scores, at local and global levels, and a visual representation of the behavior of the analysed algorithm through quality histograms.
'''EvaLTex'''
 
   
 
== Teaching ==
 
== Teaching ==

Revision as of 16:41, 18 November 2016

  • Name: Ana Stefania Calarasanu
  • Email: calarasanu@lrde.epita.fr
  • Thesis subject: Improvement of a text detection chain and the proposition of a new evaluation protocol for text detection algorithms.
  • Ecole doctorale : EDITE
  • Supervisors: Jonathan Fabrizio(LRDE) and Severine Dubuisson(ISIR)
  • LinkedIn: https://fr.linkedin.com/in/ana-stefania-calarasanu-3b4b4a52

Research interests

  • Object detection
  • Feature extraction
  • Machine learning
  • Performance measurements
  • Document image analysis

Projects

EvaLTex (Evaluating Text Localization) is a unified evaluation framework used to measure the performance of text detection and text segmentation algorithms. It takes as input text objects represented either by rectangle coordinates or by irregular masks. The output consists of a set of scores, at local and global levels, and a visual representation of the behavior of the analysed algorithm through quality histograms.

Teaching

Theory of Rational Languages

  • TD & TP to 2nd-year students at EPITA (prepa)

Algorithms

  • TD & TP to 3rd-year students at EPITA (ing1)

Publications

Conference papers

 TitlePublished inDate
From text detection to text segmentation: a unified evaluation schemeFrom text detection to text segmentation: a unified evaluation schemeProceedings of the 2nd International Workshop on Robust Reading Conference (IWRR-ECCV)1 October 2016
Towards the rectification of highly distorted textsTowards the rectification of highly distorted textsProceedings of the 11th International Conference on Computer Vision Theory and Applications (VISAPP)1 February 2016
Using histogram representation and Earth Mover's Distance as an evaluation tool for text detectionUsing histogram representation and Earth Mover's Distance as an evaluation tool for text detectionProceedings of the 13th IAPR International Conference on Document Analysis and Recognition (ICDAR)1 August 2015

Journal Papers

 AuthorTitlePublished inDate
TextCatcher: A method to detect curved and challenging text in natural scenesJonathan Fabrizio
Myriam Robert-Seidowsky
Séverine Dubuisson
Stefania Calarasanu
Raphaël Boissel
TextCatcher: A method to detect curved and challenging text in natural scenesInternational Journal on Document Analysis and Recognition8 April 2016
What is a good evaluation protocol for text localization systems? Concerns, arguments, comparisons and solutionsStefania Calarasanu
Jonathan Fabrizio
Séverine Dubuisson
What is a good evaluation protocol for text localization systems? Concerns, arguments, comparisons and solutionsImage and Vision Computing1 February 2016


Thesis

 TitlePublished inDate
Improvement of a text detection chain and the proposition of a new evaluation protocol for text detection algorithmsImprovement of a text detection chain and the proposition of a new evaluation protocol for text detection algorithms1 December 2015